Palindromic sequences that adopt hairpin and cruciform structures are a potent source of chromosomal breakage and rearrangements, and play a significant role in the pathogenesis of diseases. In humans, for example, palindromes have been found at chromosomal breakpoints of non-recurrent and recurrent translocations that can cause Emanuel syndrome. In addition, palindrome-mediated chromosomal rearrangements cause several types of ???? thalassemia, X-linked congenital hypertrichosis syndrome and hereditary renal cell carcinoma. Finally, palindromes are implicated in the amplification of genes that promote tumorigenesis in colon and breast cancer, medulloblastoma and lymphoma. Despite the critical impact of palindromes on genome maintenance and disease, how these repeats cause chromosome breakage and rearrangements in eukaryotic cells is largely unknown. The overall objective of this research is to elucidate the mechanisms of chromosomal breakage at palindromic sequences in yeast, Saccharomyces cerevisiae. The central hypothesis of the proposal is that chromosomal fragility at palindromic sequences is caused by multiple nucleases at different stages of the cell cycle and is dependent on the type of secondary structure formed. Our preliminary data indicate that there are three distinct pathways by which secondary structures can initiate double- strand break (DSB) formation and promote chromosomal instability. One pathway involves the Mre11/Rad50/Xrs2 (MRX) complex and Sae2, one depends on the structure-specific Mus81/Mms4 nuclease, and one involves an unknown nuclease. In Aim 1, we will determine factors required for DSB formation by the MRX complex and Sae2. We have found that perfect palindromes but not quasi-palindromes are sites for DSB formation by MRX/Sae2 and will test the hypothesis that nuclease attack occurs on hairpin structures formed by perfect palindromes during S-phase. Aim 2 will identify parameters of palindromic sequences that determine the specificity of secondary-structure attack by the Mus81/Mms4 nuclease. We have found that breakage at a perfect palindrome composed of actively transcribed genes is partially dependent on the Mus81/Mms4 nuclease. This nuclease does not make breaks at two other nontranscribed palindromes. We will determine the role of transcription in Mus81/Mms4 attack and will define parameters of palindromes required for such targeting. Aim 3 will identify the protein(s) responsible for MRX/Sae2- and Mus81/Mms4-independent breakage of palindromes. The hypothesis that there is another pathway for palindrome fragility that is controlled by the Lsm2-8 complex and the Cdc28 kinase will be tested. We hypothesize that this pathway involves an unknown cruciform resolvase that operates during the G2 stage of the cell cycle. The proposed research is innovative because it utilizes a unique set of sensitive tools that will allow the identification of all nucleases contributing to palindrome fragility and will determine their cleavage specificity. This proposal is significant because it will elucidate the poorly- defined mechanisms that generate chromosomal aberrations that underlie human diseases.